Lawrence Berkeley National Laboratory Lawrence Berkeley National Laboratory

Title The Isotope Effect in the Decomposition of Oxalic

Permalink https://escholarship.org/uc/item/1s96t1nf

Authors Fry, Arthur Calvin, Melvin

Publication Date 1952-07-03

eScholarship.org Powered by the California Digital Library University of California ------.

TWO-WEEK LOAN COPY

This is a Library Circulating Copy which may be borrowed for tloo weeks. For a personal retention copy, call Tech. Info. Diuision, Ext. 5545 UCRk.1565 Revised Unclassified-Chedstry Distribrrtf on

Radiation Laboratory

@ontrae% Mo, W-7405-eng-48

THE ISOTOPE EFFECT IN THE DECOI4l?OSITION OF OXALIC ACID

Az%hm Fry and Melvin Calvin July 3, 1952

Berkeley, California Rsdiatf on Laboratory and Department of Chemistry, University of California, Berkeley

ABSTRACT

The ($3 and & isotope effects in thl decomposition of oxalic acid have been studied simd&aneously at two different temperatures. In each case the 1 C 4' effect is approdmataly double the $3 effect, and both are in reasonable accord with theoretical cal~ulatfonsbased on a simple postulated mechanism for the reaction,

FOP publimtion fn The Journa 1 of Physical Chemistry

(1) The work described in this paper was sponsored by the U,S. Atomic Energy Comni ssfon.

(2) This paps was abstracted from pqPt of the thesis submitted by Arthur Fry to the Graduate Division of the Untversity of California in ~rtialfulfillment of the requirements for the Ph,D. degree, June 1951.

(3) Present address: Department of Chsmistry, University of Arkansas, F'ayettc- vf 115, ti~kansas, UCRL 1565

bs Arthur I??$and Melvin Calvin

Radiation Laboratory and Department of Chemistry, University of California, Berkeley

(1) The work described in this paper was sponsored by the U.S. Atomic Energy Commis sion.

(2) This paper was abstracted from part of the thesis submitted by Arthur Fry to the Graduate Division of the University of California in partial fulfillment of the requirements for the Ph.D. degree, June 1951.

(31 Present address : Department of Chemistry, University of Ar- kansas, Fa yetteville, Arkansas, ,

Lindsay, McElcheran a nd hod el have studield the c13 isotope effect

(4) J,G. Lindsay, D.E. McElcheran and HOG4 Thode, J. Chem. Wsna JJ, 589 (1949). in the decomposition of oxalic acid in concentrated sulfuric acid at 100%.

The results were reported as ratios of rate constants for the equations? Y The ratio kdk3r mhere C indicates (93, was found to be 1.032, while

kl/(k2 + kg) was 1.034.

Similar experiments have now been carried out using oxalic acid-1,2-& 2 , and both the d-3 and the cU isotope effects have been determined on the same samples, In addition, the decompositions have been carried out at two differ- ent temperatures, and the differences in energy and entropy of activation for decomposition by equations (2) and (3) have been calculated for both $3 and cG.

Preparation of Oxalic ~cid-l,2-~Pr - The oxslic acid-1,2-~~wed

was prepared by the reaction between c arbon dioxide-CU and potassium on sand at

360° followed by extensive purlfloat ion. Carbon dioxide generated PE~OUO

from 3,318 g. of 'barium carbonate (2,75 pe) with hot concentrated sulfuric acid was

dried and stored in a bulb for later use. A sample of sand was boiled with re~5.a and thoroughly washed and dried in order to remove metallic and organic

impurities. A 50 cc. round bottomed flask with a neck about four inches long was

filled about half fill with the purified sand. To this was added 2.4 g. of freshly cut ether-washed potassium, and the flask was connected to the vacuum - 5- UCRL- 1565

line by a tygon tube to facilitate shaking. After evacuation, the flask was

heated in a Woods metal bath to about 360°C, and shaken vigorously to spread

the molten potassium over the surface of the sand. When the potassium melted

a considerable quantity of volatile material mas released which was pumped off,

After the evacuation was complete, the temperature was maintained at 360'~~and

the carbon dioxide-$+ was allowed to flow into the reaction vessel. The re- action proceeded rapidly and was facilitated by vigorous shaking and tapping

on a eorck ring, When the reaction was complete as shown by the drop in car-

bon dioxide pressure, the flask was allowed to cool and the excess potassium

was cautiously decomposed by adding water. No trouble with fire was encountered

in this deeomposition, probably because the hydrogen quickly swept all the ow-

gen out of the flask; however, as a precaution, a vessel of liquid nitrogen into ? which the reaction vessel could be dipped was always kept handy in case the decom-

position became too violent, The sand and some black carbonaceous material formed

in the reaction was removed by filtration, and excess was added, The

solution was boiled to remove dissolved carbon dioxide which was recovered- as

barium carbonate of approximately the same specific activity as the starting ma-

terial, Wt, barium carbonate recovered = 0,467 g. (u,l$).The black carbonaceous material above could also be burned and the carbon dioxide recovered as barium car- bonate of high specific activity. Excess calcium chloride solution vias then added to the acetic acid solution, and after digestion, the dirty grey calcium precipitate was removed by centrifugation, The precipitate was dissolved I by adding a slight excess of concentrated hydrochloric acid and warming, This solution was then continuously ether extracted for three hours, The ether solu- tion was evaporated.over a small amount of water and the purification cycle was repeated starting with the calcium 2 -3~a?1-sc2?itatfolze The fhlaqueous

solution was concentrated to about 10 ec , 3112 fllte,-edr and the remaining

water and hydrochloric acid were fyophilieed off, lea---,.ng a pure white pro-

duct of anhydrous oxalic aeit, wt, = 0,3023 g,, 39,s :a*:lc., %: xb+r rms the

yields ranged from 4.5-55%, with an additional 15-20% of the activf ty befag 1-3-

covered as carbon dioxide a While no e,utPensive investigation of the reaction conditions was carried out, neither the amount of sand, the amount of potassium

nor the exact temperature appeared to be at all critical, The yields could

probably be improved somewhat by using freshly distilled potassium. Care was

required in concentrating the final oxalie aeid solution, shoe the acid will

sublime away quite rapidly if evaporated to dryness on a steam bath with a

stream of air passing over the surface. The lyophilization procedure was very satisfactory, The ogaL5.0 aeid was cheeked for radioactive impurities by

paper strip chromatography using a mixture of t-amyl and benzene sa-

turated with 3 1 hydrochloric acid ae the solvent, No radioactive spots were found except for the oxalic acid epot, and an upper limit of 0,5$ waB set for the amount of radioactive impurity, The purity of an inactive sample prepared by the above procedure was checsked by with standard base and stan- dard pemngamte. A sample of 0,1927 g, of the oxalia acid required 4J3 ce.,

1,000 2 sodium hydroxide, giving an equivalent weight of &9 (theory = 45,0), and 21.50 ce, 002000 ;potassium pemnanganate, giving an equivalent weight of

89,8 (theory = 90,0), For use in the decomposition experiments, 0,2230 ge of the oxalic acid-1,2-~? prepared above was diluted with 14.77 go of purified inaetive oxalic acid by dissolving the two samples together, filtering and lyophilizbg off the water, Base and pemanganate titrations on the product gave theoretical values for the equivalent weight. The inactive oxalie aeid was purified by recrystalPfzing Eakerfs C,P, oxalic acid two times from water, In each recrystallization only about half the product was recovered, with both

the first and last quarter of the crystals being discarded. Preparation of~100%Sulfuric Acido - The 100% sulf'uric acid used was prepared by adding 100 cc, of 30% fuming sulfuric to 150 cc. of ordinary concen-

trated sulfuric acid in a glass stoppered flask. The solution was shaken vigorously

and allowed to stand overnight, A sample of 1 cc., wt, = 1,82& go2 was diluted

with water and upon with standard 1 sodi-urn hydorxide required 36,92 cc.

for neutralization, giving a molecular weight of 97 -5 (99.4% of theory). This

same sulfuric acid stock was used in all the decomposition experiments.

Decomposition Experiments. - The decompositions mere carried out using 99bLI.%Sulfuric acid in the apparatus shown in Figure 1, Approximately one gram samples of oxalic acid were placed in the decomposition chamber, Be The apparatus was assembled, using the 99,4;& sulfuric acid to Lubricate the joints and stop- cocks. Ten cc. of the 99,4$ sulfuric acid was pipetted into the preheater, 6, and dry carbon dioxide-free oxygen was allowed to sweep through the system for about one how, The gas from the outlet of the assembly shown in Figure 1 was conducted, in turn, through a spiral trap cooled by a dry ice-fsopropy2 alcohol mixture, a spiral bubbler filled with 2 sodium hydroxide to remove the carbon dioxide, a packed furnace at d7000 C to burn the to carbon dioxide, and another sodium hydroxide filled bubbler to absorb this car- bon dioxide. Tihen the oxygen had been sweeping through the apparatus for about an hour, the spiral bubblers were filled with 2 sodium hydroxide,and the liquid in flask A was heated to reflux by means of a Glascol mantle, The vapor of the boiling liquid served as a constant temperature bath to heat the sulfiric acid and oxalic acid to the desired reaction temperature, The boiling point of the liquid in flask A determined the tempemturt- ,- f -~:,;;lzlbim0 >GC^,>~-~~FC? 6

carbon tetrachloride were chosen to give temperatu~es 133,0"6 and 83,1%,

respectively. When the liquid in flask k had beTr. r rx ,- L! -: -sr,~~slyi'x-

approximately one-half ho~rp,the stopcock was opened and %La ,uid-., i*Ld4

was allowed to flow onto the oxalie acid, At 103,~"~the reaction was very

vigorous and care had te be exercised %nadding the sulfuric acid t.,: *toid

frothing, In all cases the mlfuri~aeie was added as rapidly as possible,

The oxygen sweep aemed to stir the oxalie acid-sulfuric acid mixture. The

sweep was continued until gas bubbles were no longer observed in the sulfuric

acid solution, and t4hen for additional hour and one-half, The carbonate in the

two bubblers was collected and weighed as barium carbonate, The yields of

barfun carbonate from the carbon dioxide and carbon monoxide were quanbita-

tive, and the blanks were negligible, Isotopic Composition Measurements, - The C~ activity measurements were made using an ionization chamber and vibrating reed electrometer connected

to a Brown recorder, Tile ionizatfon chambers were filled te a standard pressure

with carbon dioxide generated in a vacuum system from the barim carbonate

samples with concentra%ed sulfuric acid, A sample of this same earbon dioxide

was stored io a bulb fez. later mass spectrometric analysis, The samples from

the carbon moroxide and carbon dioxide from a given run were measured in kame-

dfate succession 4n %he same ionization chamber and on the same instment in

order to reduce to a minimum any variations in the procedure and instrumen"es,

Several independent measurements were made on each sample from each decompo-

sition, starting wi$h the barium carbonate in each case, The c13 measurements 6 were made on a Consolidated Type 21-102 mass

(6) Thanks a re due to Dr, A. Newton and Dr, L, TaLamn af this Yabaratory for these measurements,

-- - -

spectrometer, using a mabnetic sweep to scan the peaks at m/q = 4.4 and 45, Sr- veral independent scannings of each sample were made. In several cases, the entire low m/q mass spectrum was taken using a voltage scan, and no peaks were found except those of the normal carbon dioxide pattern, Samples of normal carbon dioxide were also run during the sample determinations to check the re-

producibility of the mass spectrometer, The m/q = 45 peak height was corrected

for the 017 oontent of the carbon dioxide.

The cUo2 specific activities and C 1302 mole fractions obtained in the various runs are shown in Table I, In all cases, the values given are averages

of several measurements, and the indicated errors are average deviations sf

these measurement so

The (213 measurements were made at two different times, runs 1-4 and normal carbon dioxide 1 and 2 (carbon dioxide from dry ice) being measured at one time and runs 5-8 and normal carbon dioxfde 3 and 4 (same stock sample of carbon dioxide) being measured somewhat later. Since in the subsequent ,calcu- lations, only the relative cI3o2 fractions are used, the variation in the nor- mal carbon dioxide value between the two times is not serious since the differ- ence is reflected in the values from both the carbon monoxide and carbon dioxideo

The GUmeasurements were made using diqferent instruments and at different times Table 1

c%~ Specific Activities and C 33O2 lo;; R4act-.rns fr.c the DE mqxsr--..ir: ;t

Speclfic ~ctivit~~^-- T emp , Drift Fate, ~olts/min, x 103 105 CX), Fraction CO Fraction C02 Fraction CO Fraction

Nsrmal 02 B

* Corrected for presence of ~~0~70~~~

* Actual specific activity = ~30dis,/min./mg. BaCO3 which probably accounts for the variations in specific activity from run

to run, However, the carbon dioxide and carbon monoxide fractions from the same run were measured in immediate succession on the same instrument, to

reduce this variation as much as possible within a given run, In the following 'calculations only the monolabeled oxalie acid is con- sidered although the oxalio acid actually used is called oxalic acid-1,2-9,

since by its method of preparation both carbon atoms would be labeled if pure c%~ were used. Actually, the number of dilabeled molecules was calculated to

be a very negligible fraction of the total, rn Referring to equations (I), (2) and (3) and assuming that the reaction

is first order with respect to oxalic acid, the specific activities or mole

fractions of the carbon dioxide and carbon monoxide are given by the follow- ing equations:

and

Combining the above equations gives equation (4) : It should be coted that k2/k3 calculated in thei j n:3:1ner 15 ir,:':,+f;x-wt 32 5;~~- and hence cf tho, anoxat of reaetSan, while ccxnpr;.: - ;fie sijb;iZil ~:tl?'~b, of the carbon dfox2de or carbon monoxide to that of trA~:--LIE a225 : --'- 0

For ease of tabulation and convenience of referel-: - 'TV- :)e-. -I.+":<'A ,t? isots-e affect nzy be defined as 100 (k2/k3 - 1), The ~ercenta~drsG.2~~ - effect values salculated by equation (4) from tne data in TabPe I are given in Table II, The data of Lindsay, MeElcheran and Thode4 have been recalm- lated in this mamer, and show a muh narrower spmad than using their method of calcxlation, although the average values are the same, These recalculated vabes are also shown in Table 11,

Table I1

G3 and (1uIsotope Effect in the Decomposition of Oxalio icid Discussion

The $3 isotope effect values shown in Table I1 are thus in very satis- 4 factory agreement with the results obtained by Lindsay, McElcheran and Thode . The cU effect is approximately double the c'~ effect at each temperature.

The differences in energy and entropy of activation caused by the iso- 7 topic substitution can be calculated from the theory of absolute rates,

('(1 So Glasstone, K, Ja Laidler and H, Eyring, "The Theory of Rate Pro- eessest\ lst, ed., McGraw-Hill Book Company, New York, Nevi York (1941).

------

For equation (21, and temperature TI,

and for equetion (3), and TI,

Combining,

In a similar manner an equation is set up for T2, and when this equation is corn- bfned with (5), equation (6) results, From the data observed in Table 11, (bI$12 - b@3* can be calculated by equation (6), and this value can be substituted back into equation (5), and (sA#~- h.Sf3) om be calculated. The oelculated values are shovin in Table 111.

Ta.ble I11

Energy and Entropy of BetivatSon Difference between Equations (2) and (33

The presence of the isotope effect measured here allows some rather specific conclus5.ons to be drawn concerning the mechanism of the decompo- sition,' A5 complete reaction, all of the carbon-carbon bonds in all the

(8) The authors wf sh to express their thanks to Professor Richard Powell for a fntftfil dlssussion of the mechanism of the decomposition, moPeoules are broken, so the %sotope effect obsemed cannot be due to fso- topfc discrimination in the rate of rupture of the carbon-carbon bondso Therefore, if the observed isotopic fractionation is due to an isotope effect in the rate detemining step of the reaction, the rate determining step mast be the breaking of a carbon-oxygen bond, since the association of hydrogen ~5thoxygen is eoimnly considered to be rapid and reversible,

However, the observed isotope r0"ect nlight also be due to some equilibrium step, An fsokope effect i:, conceivable yn rapid steps fol- i-ming: the rate cleternining step only under very special ~ircumstances,

These mquire a s~+ietrical starting molecule and either a symnetrical pro- duct of the rate detedning step or products which are rapidly intercon-

~9rtablecolnpared to the irreversible following reactions, in which the lsotopa selection is made. If an equilibrium process occurs before the rate deterdining step, an isotope effect in the equtlibrium could multiply an isotope effect in the rate determining step, aefd a s shown fr equatf on ('71,

IA simple ca1culstionY the tyuilibrfm constant, R, may be mddc using the equa- 0 10 tions derived bj Efgeleiaar, and Uayer' o~ by Urey Fw Eaek of complete ckatt., we

eefvable %ha%in oth~reases tho effect would be in th2 spposlte 2Srecf%on, UCRL-1565 Revised

carbon bond f s gcatly ~i%!t%l2d leaving nearly free mol%cules of carbon dioxide snd. 8 -1 COOH, The carbor! monoxidc i~ assigned its acce2ted frequency of 21(0 cm. and the e; carbonyl and hydroxyl frequenciss in COOH are assumed to be unchanged. Ths force eon- stant for tli? C-OH bond which is broken is set equal to zsro, Tkc: ratio k2/k3 is 11 then ccleulsted using equation (8), as derive2 by Bigeleis5n

9 The function ~(u)is defined by Bigeleisen and Mayer who have tabulated values of

G(U) 9s a function of u, The reaction coordinate reduced mass, p., is assumed to be the reduced mass of the atoms forming the bond being broken, The symbol # re-

fs~to the activcted complex,

The rate ratio, k2/kj, calcul~tedin this manner, when multiplied by the equili-

brim constants calcultts? above, give overall isotope effects of 6,0$ for CU sub-

stitution 3nd 3.15 for c13 substitution at 100°. These compare to the valoes of

5.5% 2nd 2,7% mzasured experimentally. The corresponding cslculated values at $80 - 1.2 are 6.3% and 3.4$ compared to the c::perfmental values of 6.75 and 3.1c39,. . The agree-

We hsve recently learned, by private communicztion, of experiments in two othrr places (Research Chemistry Division, Atomic Energy Project, Idn,tional Ele~~enrchCouncil Chalk Rivs, Canada; and Department of Chsmistsy, Univer- sity of 1llfnolsj in which the isotope effect in s chemical reaction was de- termined for both $3 and c14 simulttneously on the same reaction. +n both of thrse oasis the mported CU effect is much more than trice the ~'3effect. The results for the decompositfon of mesitoic acid giwn by Stevens, Pepper snd Lounsbury (~anada),J, Chern, Phys., 20, 192 (1952) are as follows: k12/k13 1.03P t.003; k12/k14, 1.131 +_ .005. Here the C~/CI~ratio is almost three. The T -:ults reported by Yanhich, Stivers and Nystrom (Illinois) at the study sf isotope eff~ctsi~ chemical reactions should enable us to elucidsta ciLi

of the intimate dateils of the ~!eehanisrncf the reaction. Whoc the propc~en;;iliSrSa

tw3 chemically f2antica2 g~oups, such as oxalic or , is a f3ri;unate em, sfnee n5t isot~pesffect~ @re abserved at complete rhsction, thus making it unnecesaiar:.

Lo study the isotopie composition cf the products as a function cf the amount of

I-%acb,Em.

The pres5nee of aE fsotop effect dgcs not unccpfvoca1l~-ests3iish t'13 rwehcnis-i. of the deconpositien, '?:t it does off sr vsx-y stmng eviiiance Plhct. Lhe z-dptul-e or" ti:.. cs-bsr-earban bond is not r~te determining, and, consequentl;r, thzit tLe rate detemin- ing step in this medium is the rupture of the carbon-oxygen bond. hct~aluthe rvptixe of the carbon-oxygsn bond and of the carbon-carbon bond is prcbekfy all part of 6 conce~tedrsactfon with the loss of hydroxy1 (or water) from one carboxyl or

Path A 4- Path B 9

Path A 5. s seen Lo be less likely than Path B, since we obseme an enrich- ment of labeied atoms in the carbon d-ioxidt, 3N - 6 the tern G(U. ;by2 ;in ~~c~uatioa(7) is alliri j~ identisa1.l.y zeroI thus - - A .L affordlcg a considerable simplification in the caiculations for the rzte deter-